One of the key pillars of the magnificent economic temple Mr. Ben Bernanke has constructed in the clouds (where his head is) is the notion that he can predict inflation, and stamp it out before it gets to be a problem. Since controlling inflation is at least half of the job for which we citizens have ostensibly hired Ben, I've searched high and low for a set of inflation predictions he has made personally as evidence of his skill in this area. Finding none, I turned briefly to the past projections made by the FOMC as a group. These are not exactly stellar (perhaps the topic of another article), but there isn't enough data there to analyze.
How, then, does the FOMC determine the general direction of the future value of a dollar? Quite simply, it asks all the rest of us, specifically bond investors, what we think! Ben, like the Maestro before him, is afflicted with the touching belief that the madness of the bond investor crowd (made up of millions of large and small investors making decisions based on all sorts of economic and non-economic, rational and emotional, factors) will magically distill down to a net-net correct view of future inflation! To quote the Fed's December statement:
In determining how long to maintain a highly accommodative stance of monetary policy, the Committee will also consider… indicators of inflation pressures and inflation expectations, and readings on financial developments.
Being an unapologetic fan of free markets, I am sympathetic to the notion that disparate opinions about value, if freely expressed in markets, often result in prices close to "fair value" (whatever that is). But investors have embedded biases, too, and the more a particular market is dominated by non-economic decision-makers (as is clearly the case in Treasury bond land, where Ben is, incredibly, warping the market for the very instruments on whose prices he relies to frame interest rate decisions), the less accurate its prices will be.
But we don't need to get deep into the efficient market debate in order to determine whether bond investors can accurately predict inflation. We know they can't! How do we know? We can just examine their past track record.
TIPS prices (TIP) (STPZ) are the most direct way for bond investors to guess at future inflation. The Fed has even constructed an index to track their guesses: the "Fed's Five-year Forward Breakeven Inflation Rate" (FED5YEAR Index for those with Bloomberg). Amazingly enough, this index is almost completely uncorrelated with subsequent inflation. (I used the CPURNSA Index for actual inflation.)
Not exactly a great track record for TIPS investors, eh? But let's cut them, and the Fed, some slack. After all, TIPS have been around for less than 15 years, and we can hardly call that a fair trial.
Unfortunately for Ben, however, the picture really doesn't improve at all when you look further back in time, and attempt to rely on "regular" Treasury bond prices to predict future inflation. For example, let's examine the yields on 5-year Treasury bonds (IEI), for which I have access to data back to 1962 (GT5 Govt on Bloomberg). Here is 5-year inflation (y-axis) plotted against the 5-year Treasury bond yield at the outset of that period:
Not only is there a very low correlation, but the bond yield has historically provided an especially poor prediction at extremes - just when one would hope to rely on it the most. To quantify this, note that the highest decile of Treasury bond yield observations, with a median of 12.7%, were followed by 5-year inflation at a median rate of only 3.6%. By contrast, the highest decile of 5-year inflation readings, with a median of 9.1%, were preceded by 5-year bond yields of only 7.5% (only slightly above the median bond yield for the period).
And yet conventional economic "wisdom" - guided by a single British gambler/economist whose theories have failed multiple analytical tests - continues to say we should set monetary policy based on inflation expectations as measured in the bond market. Shocking!
As a side note, we might wonder: What are bond investors are doing if they are not predicting inflation? The answer is that they are reacting to it, just like "less intelligent" traders in other markets. For example, the correlation in the above chart (square root of .06) is 24%, but when you regress historical 5-year inflation against the 5-year Treasury bond yield at the end of that period, the correlation rises to 81%. Using various lags, the point of maximum correlation comes by regressing 5-year inflation ending as of a certain date against the 5-year Treasury bond yield five months after that date.
None of this is surprising to me. Cognitive dissonance, however, arises from the idea that the Keynesians at the Fed have either never conducted such simple tests of their theories (leading to the conclusion that they are not the academic geeks I thought they were), or have ignored the conclusions of these tests (leading to the conclusion that they are not interested in the truth). Either way, it's time for the Federal Reserve to stop trying to control the time value of money. Until they do, gold (GLD) (IAU) may be a good long-term investment.